The reason that this approach isn't faster is because each task has to do the "is particle x in cell y" computation for all 16 possible values of y in each task. Because of this, even though there are 16 parallel tasks, they each have to do 16 times more work so the result is no faster than the sequential algorithm.
The reason that this approach isn't faster is because each task has to do the "is particle x in cell y" computation for all 16 possible values of y in each task. Because of this, even though there are 16 parallel tasks, they each have to do 16 times more work so the result is no faster than the sequential algorithm.
This comment was marked helpful 0 times.